Pose Lab runs on your own machine — the client starts it, so there is no endpoint to ping. 200 installs a week from pypi. Last commit 31 Oct 2026.
Measured spatial answers for posing first-person arms and a rifle in Blender
We read the source, 1 d ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
self.proc = subprocess.Popen([find_blender(), "-b", "--factory-startup", "-P", str(WORKER), "--", str(port), rigs, str(out)],
path = os.path.join(self.folder, f)
centre_v = centre + drop
cam.location = cam.location + drop
env = dict(os.environ, POSELAB_TOKEN=self.token)
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add poselab-mcp -- uvx poselab-mcp
{
"mcpServers": {
"poselab-mcp": {
"command": "uvx",
"args": [
"poselab-mcp"
]
}
}
}
[mcp_servers.poselab-mcp]
command = "uvx"
args = ["poselab-mcp"]
{
"mcpServers": {
"poselab-mcp": {
"command": "uvx",
"args": [
"poselab-mcp"
]
}
}
}
{
"servers": {
"poselab-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"poselab-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
POSELAB_BLENDER (Blender's executable, if it is not on the PATH or in the usual install folder), POSELAB_RIGS (A rigs.json describing your own FBX rigs; the built-in sample rig needs none), POSELAB_OUT (The only folder Pose Lab writes to (renders, the worker's log); defaults to ~/.poselab).
Find and fix first-person weapon bugs behind a gate: your game's probe measures, the Gate judges
Local-first persistent memory for AI agents, with reasoning and formal verification.
Bounded footwear fit checks for a specific shoe, person, and intended use.
Measured latency, time to first token and uptime for ~45 AI inference APIs, by region.
Persistent local memory for MCP-compatible AI coding agents, stored per repository in SQLite.
Local-first personal memory and model server for trusted MCP agents on macOS.
Audits how AI answer engines read a page and grades it A+ to F across four measured layers.
Checks GLB assets against a spec without Blender and embeds the measured proof inside the GLB.
Answers built from our own checks of this server.